trcclub / autonomous-forest-surveillance-safety-system-using-opencv Goto Github PK

This project forked from mohit9949/autonomous-forest-surveillance-safety-system-using-opencv

The current forest surveillance systems methods consume a lot of resources and are less efficient, not reliable and require a constant human presence whose tasks can be easily automated using new technology. To solve these problems we propose an autonomous surveillance system which uses object detection to identify specified animals. It is capable of monitoring forest fires, intruders, wildlife etc, all at once and alerts the concerned officials immediately and precisely. It has a hybrid object detection system using HAAR and Backpropagation neural network algorithms which can be used to train and detect animals and predict from the data obtained respectively. This helps in detecting various unwanted visitors, dangerous animals, or restricted tools into the forest. The system can not only store the video feed but can also determine population , track a specific animal or human and sends the pictures to your email directly along with real-time video monitoring via the internet which allows the users to monitor from anywhere in the world and sends instant alerts to your phone via an SMS even in remote areas in case of emergencies, and it stores all the data in a repository. We can control the system using a windows app which allows us to select which animals to be detected by the camera modules and their alert levels along with other settings and also provides a detailed analysis on various things like forest fires, animal population, trespassed areas etc, to users in simple charts. It is a smart, automatic, modular system which is cheap and easily expandable.